Citation

Stattin H, Magnusson D. Br. J. Criminol. 1991; 31(4): 327-346.

Copyright

(Copyright © 1991, Centre for Crime and Justice Studies, Publisher Oxford University Press)

DOI

unavailable

PMID

unavailable

Abstract

The issue of stability and change in criminal behaviour was investigated using data for registered crime obtained in a Swedish prospective, longitudinal research project. Criminal activity among 709 males was investigated for three time periods: childhood (up to age 14), adolescence (from age 15 to age 20), and early adulthood (age 21-30). The Spearman correlation coefficient, reflecting the stability of registered criminal activity, was 0.40 between childhood and adolescence, 0.34 between childhood and early adulthood, and 0.38 between adolescence and early adulthood. Little specialization of criminal activity over time was revealed. Rather, a diversified crime pattern in early adulthood tended to be preceded by a spread of offences over different types of crime in adolescence. The small group of subjects who had committed offences below age 15, during adolescence, and also in early adulthood were responsible for most of the crime recorded for the total research population. These persistent offenders constituted 5.4 per cent of the total number of subjects, were recorded to have 41 per cent of all registered convictions, and participated in 62 per cent of all the registered crime occasions. Their over-representation with regard to offences committed was relatively uniform across different types of crime, both in adolescence and in early adulthood. Overall, acquiring a record for early delinquency was quite strongly prognostic of future crime. More than three out of four of the subjects who appeared on the police registers before the age of criminal responsibility also showed up there later. The relevance of early onset of criminality for criminal behaviour in the long term is discussed.